7 Days 7 Machine Learning & Python Projects From Scratch

Set up a Google Colab notebook for phishing website prediction with a random forest, import pandas and scikit-learn packages, and prepare data using train_test_split, accuracy score, and standard scalar.

Load a csv dataset in Colab with pandas read_csv, inspect with head and tail, and train a random forest classifier to predict fake websites using the class label.

Learn to split a data set into features and the target variable by dropping the id and class label, creating x and y for classification. Next, train with random forest.

Split the data into 80/20 training and testing sets, scale features with standard scaler, and train a random forest classifier with 100 estimators and a 42 random state.

Complete your project by predicting outputs with the random forest classifier, evaluating accuracy on the test set, and concluding with a 98% score for fake website detection.

Build a cryptography project that performs encryption and decryption using the Fernet symmetric method, generate the key with Fernet, and apply the key for secure encryption and decryption.

Set up a Google Colab notebook, name the project, and import packages to implement Fernet symmetric encryption and decryption using a single key; install cryptography and import Fernet and OS.

In this class, students write a function to generate a Fernet key, save it to a file named encrypted key, and print the file name for encryption and decryption.

Load the generated key for encryption and decryption by implementing a load key function that reads the key from a file named encrypted.key and opens it in read mode.

encrypt a file with the fernet symmetric method by loading the key, creating a fernet object, and writing encrypted data to a .encrypted file.

Learn to decrypt files with the Fernet symmetric method by loading the key, reading encrypted data, applying Fernet decrypt, and writing the decrypted output to the decrypted file.

Generate a key and encrypt and decrypt files with the fernet symmetric method by implementing a main function that offers choices for key generation, encryption, and decryption.

Develop an e-commerce FAQ bot from scratch using Python and NLP in Google Colab, building a question–answer dataset, vectorizing text with TF-IDF, and calculating cosine similarity to fetch responses.

Set up a simple file transfer between client and server using Python sockets, with server.py and client.py, binding 0.0.0.0:50001, and a 4096 buffer to receive and write a file.

Build a fake news detector with naive bayes and Streamlit. Load csv data, vectorize titles, train a multinomial nb model, and deploy an interactive real-or-fake predictor.